Upload folder using huggingface_hub
Browse files- .gitattributes +20 -0
- README.md +143 -0
- imat-bf16-gmerged.dat +3 -0
- sfr-iterative-dpo-llama-3-8b-r-bf16.gguf +3 -0
- sfr-iterative-dpo-llama-3-8b-r-imat-IQ1_S.gguf +3 -0
- sfr-iterative-dpo-llama-3-8b-r-imat-IQ2_M.gguf +3 -0
- sfr-iterative-dpo-llama-3-8b-r-imat-IQ2_S.gguf +3 -0
- sfr-iterative-dpo-llama-3-8b-r-imat-IQ2_XS.gguf +3 -0
- sfr-iterative-dpo-llama-3-8b-r-imat-IQ2_XXS.gguf +3 -0
- sfr-iterative-dpo-llama-3-8b-r-imat-IQ3_M.gguf +3 -0
- sfr-iterative-dpo-llama-3-8b-r-imat-IQ3_S.gguf +3 -0
- sfr-iterative-dpo-llama-3-8b-r-imat-IQ3_XS.gguf +3 -0
- sfr-iterative-dpo-llama-3-8b-r-imat-IQ3_XXS.gguf +3 -0
- sfr-iterative-dpo-llama-3-8b-r-imat-IQ4_NL.gguf +3 -0
- sfr-iterative-dpo-llama-3-8b-r-imat-IQ4_XS.gguf +3 -0
- sfr-iterative-dpo-llama-3-8b-r-imat-Q4_0.gguf +3 -0
- sfr-iterative-dpo-llama-3-8b-r-imat-Q4_K_M.gguf +3 -0
- sfr-iterative-dpo-llama-3-8b-r-imat-Q4_K_S.gguf +3 -0
- sfr-iterative-dpo-llama-3-8b-r-imat-Q5_K_M.gguf +3 -0
- sfr-iterative-dpo-llama-3-8b-r-imat-Q5_K_S.gguf +3 -0
- sfr-iterative-dpo-llama-3-8b-r-imat-Q6_K.gguf +3 -0
- sfr-iterative-dpo-llama-3-8b-r-imat-Q8_0.gguf +3 -0
.gitattributes
CHANGED
@@ -33,3 +33,23 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
36 |
+
imat-bf16-gmerged.dat filter=lfs diff=lfs merge=lfs -text
|
37 |
+
sfr-iterative-dpo-llama-3-8b-r-bf16.gguf filter=lfs diff=lfs merge=lfs -text
|
38 |
+
sfr-iterative-dpo-llama-3-8b-r-imat-IQ1_S.gguf filter=lfs diff=lfs merge=lfs -text
|
39 |
+
sfr-iterative-dpo-llama-3-8b-r-imat-IQ2_M.gguf filter=lfs diff=lfs merge=lfs -text
|
40 |
+
sfr-iterative-dpo-llama-3-8b-r-imat-IQ2_S.gguf filter=lfs diff=lfs merge=lfs -text
|
41 |
+
sfr-iterative-dpo-llama-3-8b-r-imat-IQ2_XS.gguf filter=lfs diff=lfs merge=lfs -text
|
42 |
+
sfr-iterative-dpo-llama-3-8b-r-imat-IQ2_XXS.gguf filter=lfs diff=lfs merge=lfs -text
|
43 |
+
sfr-iterative-dpo-llama-3-8b-r-imat-IQ3_M.gguf filter=lfs diff=lfs merge=lfs -text
|
44 |
+
sfr-iterative-dpo-llama-3-8b-r-imat-IQ3_S.gguf filter=lfs diff=lfs merge=lfs -text
|
45 |
+
sfr-iterative-dpo-llama-3-8b-r-imat-IQ3_XS.gguf filter=lfs diff=lfs merge=lfs -text
|
46 |
+
sfr-iterative-dpo-llama-3-8b-r-imat-IQ3_XXS.gguf filter=lfs diff=lfs merge=lfs -text
|
47 |
+
sfr-iterative-dpo-llama-3-8b-r-imat-IQ4_NL.gguf filter=lfs diff=lfs merge=lfs -text
|
48 |
+
sfr-iterative-dpo-llama-3-8b-r-imat-IQ4_XS.gguf filter=lfs diff=lfs merge=lfs -text
|
49 |
+
sfr-iterative-dpo-llama-3-8b-r-imat-Q4_0.gguf filter=lfs diff=lfs merge=lfs -text
|
50 |
+
sfr-iterative-dpo-llama-3-8b-r-imat-Q4_K_M.gguf filter=lfs diff=lfs merge=lfs -text
|
51 |
+
sfr-iterative-dpo-llama-3-8b-r-imat-Q4_K_S.gguf filter=lfs diff=lfs merge=lfs -text
|
52 |
+
sfr-iterative-dpo-llama-3-8b-r-imat-Q5_K_M.gguf filter=lfs diff=lfs merge=lfs -text
|
53 |
+
sfr-iterative-dpo-llama-3-8b-r-imat-Q5_K_S.gguf filter=lfs diff=lfs merge=lfs -text
|
54 |
+
sfr-iterative-dpo-llama-3-8b-r-imat-Q6_K.gguf filter=lfs diff=lfs merge=lfs -text
|
55 |
+
sfr-iterative-dpo-llama-3-8b-r-imat-Q8_0.gguf filter=lfs diff=lfs merge=lfs -text
|
README.md
ADDED
@@ -0,0 +1,143 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: cc-by-nc-nd-3.0
|
3 |
+
pipeline_tag: text-generation
|
4 |
+
base_model: Salesforce/SFR-Iterative-DPO-LLaMA-3-8B-R
|
5 |
+
tags:
|
6 |
+
- salesforce
|
7 |
+
- llama
|
8 |
+
- llama-3
|
9 |
+
- instruct
|
10 |
+
- finetune
|
11 |
+
- gguf
|
12 |
+
- imatrix
|
13 |
+
- importance matrix
|
14 |
+
model-index:
|
15 |
+
- name: SFR-Iterative-DPO-LLaMA-3-8B-R-iMat-GGUF
|
16 |
+
results: []
|
17 |
+
---
|
18 |
+
|
19 |
+
# Quant Infos
|
20 |
+
|
21 |
+
- quants done with an importance matrix for improved quantization loss
|
22 |
+
- gguf & imatrix generated from bf16 for "optimal" accuracy loss
|
23 |
+
- Wide coverage of different gguf quant types from Q\_8\_0 down to IQ1\_S
|
24 |
+
- Quantized with [llama.cpp](https://github.com/ggerganov/llama.cpp) commit [dc685be46622a8fabfd57cfa804237c8f15679b8](https://github.com/ggerganov/llama.cpp/commit/dc685be46622a8fabfd57cfa804237c8f15679b8) (master as of 2024-05-12)
|
25 |
+
- Imatrix generated with [this](https://github.com/ggerganov/llama.cpp/discussions/5263#discussioncomment-8395384) multi-purpose dataset.
|
26 |
+
```
|
27 |
+
./imatrix -c 512 -m $model_name-f16.gguf -f $llama_cpp_path/groups_merged.txt -o $out_path/imat-f16-gmerged.dat
|
28 |
+
```
|
29 |
+
|
30 |
+
# Original Model Card:
|
31 |
+
|
32 |
+
|
33 |
+
# SFR-Iterative-DPO-Llama-3-8B-R
|
34 |
+
|
35 |
+
## Introduction
|
36 |
+
We release a state-of-the-art instruct model of its class, **SFR-Iterative-DPO-LLaMA-3-8B-R**.
|
37 |
+
On all three widely-used instruct model benchmarks: **Alpaca-Eval-V2**, **MT-Bench**, **Chat-Arena-Hard**, our model outperforms all models of similar size (e.g., LLaMA-3-8B-it), most large open-sourced models (e.g., Mixtral-8x7B-it),
|
38 |
+
and strong proprietary models (e.g., GPT-3.5-turbo-0613). The model is trained with open-sourced datasets without any additional human-/GPT4-labeling.
|
39 |
+
|
40 |
+
## Model Releases
|
41 |
+
- [SFT model](https://huggingface.co/Salesforce/SFR-SFT-LLaMA-3-8B-R)
|
42 |
+
- [Reward model](https://huggingface.co/Salesforce/SFR-RM-LLaMA-3-8B-R)
|
43 |
+
- [RLHF model](https://huggingface.co/Salesforce/SFR-Iterative-DPO-LLaMA-3-8B-R)
|
44 |
+
|
45 |
+
|
46 |
+
## Training methods
|
47 |
+
We have developed a simple and efficient online RLHF recipe for LLM instruct training. Our recipe is DPO-based and thus much cheaper and simpler to train and tune compared to PPO-based approaches.
|
48 |
+
Unlike widely-used offline DPO, the online component of our approach effectively mitigates distribution shifts during policy optimization.
|
49 |
+
For a detailed exposition, please refer to our accompanying technical report.
|
50 |
+
|
51 |
+
|
52 |
+
## Chat Benchmarks
|
53 |
+
|
54 |
+
| **Model** | **Size** | **Method** | **LC Alpaca-Eval-V2** | **MT-Bench** | **Chat-Arena-Hard** |
|
55 |
+
|-------------------------|----------|-------------------|-----------------------|--------------|---------------------|
|
56 |
+
| **Small Open-Sourced Models** | | | | | |
|
57 |
+
| Gemma-7B-it | 7B | SFT | 10.4 | 6.38 | 7.5 |
|
58 |
+
| Zephyr-7B-beta | 7B | Vanilla DPO | 13.1 | 7.34 | - |
|
59 |
+
| Mistral-7B-v0.2-it | 7B | SFT | 17.1 | 7.51 | 12.6 |
|
60 |
+
| Open-Chat-0106 | 7B | SFT | 15.6 | 7.8 | - |
|
61 |
+
| Starling-7B-beta | 7B | PPO | 25.8 | 8.12 | 23.0 |
|
62 |
+
| LLaMA-3-8B-it | 8B | RS+DPO+PPO | 22.9 | 8.16 | 20.6 |
|
63 |
+
| **Ours** | | | | | |
|
64 |
+
| Ours (SFT baseline) | 8B | SFT | 10.2 | 7.69 | 5.6 |
|
65 |
+
| Ours (DPO baseline) | 8B | Vanilla DPO | 22.5 | 8.17 | 22.4 |
|
66 |
+
| Ours (Online RLHF) | 8B | Iterative DPO | **37.2** | **8.46** | **29.1** |
|
67 |
+
| **Large Open-Sourced Models** | | | | | |
|
68 |
+
| Vicuna-33b-v1.3 | 33B | SFT | 17.6 | 7.12 | 8.6 |
|
69 |
+
| Yi-34B-Chat | 34B | SFT | 27.2 | - | 23.1 |
|
70 |
+
| Mixtral-8x7B-it | 45B* | SFT | 23.7 | 8.30 | 23.4 |
|
71 |
+
| Tulu-2-DPO-70B | 70B | Vanilla DPO | 21.2 | 7.89 | 15.0 |
|
72 |
+
| LLaMA-3-70B-it | 70B | RS+DPO+PPO | 34.4 | 8.95 | 41.1 |
|
73 |
+
| Mixtral-8x22B-it | 141B* | SFT | 30.9 | 8.66 | 36.4 |
|
74 |
+
| **Proprietary Models** | | | | | |
|
75 |
+
| GPT-3.5-turbo-1106 | - | - | 19.3 | 8.35 | 18.9 |
|
76 |
+
| GPT-3.5-turbo-0613 | - | - | 22.7 | 8.39 | 24.8 |
|
77 |
+
| GPT-4-0613 | - | - | 30.2 | 9.18 | 37.9 |
|
78 |
+
| Claude-3-Opus | - | - | 40.5 | 9.00 | 60.4 |
|
79 |
+
| GPT-4 Turbo (04/09) | - | - | 55.0 | - | 82.6 |
|
80 |
+
|
81 |
+
|
82 |
+
## Academic Benchmarks
|
83 |
+
|
84 |
+
| **Model** | **Size** | **Method** | **GSM-8K** | **MMLU** | **HumanEval** | **TruthfulQA** | **ARC** | **MBPP** |
|
85 |
+
|----------------------------|----------|-----------------|------------|----------|---------------|----------------|---------|----------|
|
86 |
+
| LLaMA-3-8B-it | 8B | RS+DPO+PPO | 79.6 | 66.0 | 61.6 | 43.9 | 59.5 | 61.1 |
|
87 |
+
| Ours (SFT baseline) | 8B | SFT | 74.2 | 64.7 | 65.2 | 53.4 | 61.4 | 62.3 |
|
88 |
+
| Ours (DPO baseline) | 8B | Vanilla DPO | 79.8 | 64.5 | 63.4 | 61.8 | 65.2 | 60.3 |
|
89 |
+
| Ours (Iterative RLHF) | 8B | Iterative DPO | 80.7 | 65.3 | 64.6 | 60.4 | 64.3 | 60.8 |
|
90 |
+
|
91 |
+
|
92 |
+
## Usage
|
93 |
+
```python
|
94 |
+
from transformers import AutoModelForCausalLM, AutoTokenizer
|
95 |
+
|
96 |
+
device = "cuda"
|
97 |
+
|
98 |
+
model = AutoModelForCausalLM.from_pretrained("Salesforce/SFR-Iterative-DPO-LLaMA-3-8B-R")
|
99 |
+
tokenizer = AutoTokenizer.from_pretrained("Salesforce/SFR-Iterative-DPO-LLaMA-3-8B-R")
|
100 |
+
|
101 |
+
messages = [
|
102 |
+
{"role": "user", "content": "I'm trying to teach myself to have nicer handwriting. Can you help?"},
|
103 |
+
]
|
104 |
+
|
105 |
+
model_inputs = tokenizer.apply_chat_template(messages, return_tensors="pt")
|
106 |
+
|
107 |
+
model_inputs = model_inputs.to(device)
|
108 |
+
model.to(device)
|
109 |
+
|
110 |
+
output_tokens = model.generate(model_inputs, max_new_tokens=1024, do_sample=True)
|
111 |
+
model_outputs = tokenizer.batch_decode(output_tokens)
|
112 |
+
print(model_outputs[0])
|
113 |
+
```
|
114 |
+
|
115 |
+
|
116 |
+
## Limitations
|
117 |
+
SFR-Iterative-DPO-LLaMA-3-8B-R is a research model developed as part of our RLHF initiative at Salesforce.
|
118 |
+
While safety and ethical considerations are integral to our alignment process,
|
119 |
+
there remains the possibility that the model could generate offensive or unethical content, particularly under adversarial conditions.
|
120 |
+
We are committed to continuous improvement in our models to minimize such risks and encourage responsible usage.
|
121 |
+
|
122 |
+
## Citation
|
123 |
+
Please cite our papers if you find our models are useful.
|
124 |
+
|
125 |
+
```bibtex
|
126 |
+
@misc{dong2024rlhf,
|
127 |
+
title={RLHF Workflow: From Reward Modeling to Online RLHF},
|
128 |
+
author={Hanze Dong and Wei Xiong and Bo Pang and Haoxiang Wang and Han Zhao and Yingbo Zhou and Nan Jiang and Doyen Sahoo and Caiming Xiong and Tong Zhang},
|
129 |
+
year={2024},
|
130 |
+
eprint={2405.07863},
|
131 |
+
archivePrefix={arXiv},
|
132 |
+
primaryClass={cs.LG}
|
133 |
+
}
|
134 |
+
|
135 |
+
@misc{xiong2024iterative,
|
136 |
+
title={Iterative Preference Learning from Human Feedback: Bridging Theory and Practice for RLHF under KL-Constraint},
|
137 |
+
author={Wei Xiong and Hanze Dong and Chenlu Ye and Ziqi Wang and Han Zhong and Heng Ji and Nan Jiang and Tong Zhang},
|
138 |
+
year={2024},
|
139 |
+
eprint={2312.11456},
|
140 |
+
archivePrefix={arXiv},
|
141 |
+
primaryClass={cs.LG}
|
142 |
+
}
|
143 |
+
```
|
imat-bf16-gmerged.dat
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ee53f56f2874d69c49737c949b2f85d0ce5eba44576dae4037eb5ceecfc6f9cd
|
3 |
+
size 4988185
|
sfr-iterative-dpo-llama-3-8b-r-bf16.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:9662fac39ebb5cd1efc15758d2a1158beead31afc4de58fbbb8f9a2e6aa90339
|
3 |
+
size 16068890912
|
sfr-iterative-dpo-llama-3-8b-r-imat-IQ1_S.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:78e90d03a961e7be9161af5a9e4b2252d53c3ee3d3f0f2c3fff600a358768142
|
3 |
+
size 2019627616
|
sfr-iterative-dpo-llama-3-8b-r-imat-IQ2_M.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:23fdb43c8f18011fa2156a3711241e7a74d70c4d1913c7995a578f52b7242e76
|
3 |
+
size 2948280928
|
sfr-iterative-dpo-llama-3-8b-r-imat-IQ2_S.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:0703b7a6fcd83badd3180052226b7ca8139c13c0563f6e2769918fd19763a492
|
3 |
+
size 2758488672
|
sfr-iterative-dpo-llama-3-8b-r-imat-IQ2_XS.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:700dcbc8b551af6d4e2638ec3eece08d10ad515d641e245a0606a3038ff7382a
|
3 |
+
size 2605781600
|
sfr-iterative-dpo-llama-3-8b-r-imat-IQ2_XXS.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:9bdaac3b6f1b8401382dd667acb36c057360c0c18e1f13f4579aaf7aabdca7ad
|
3 |
+
size 2399212128
|
sfr-iterative-dpo-llama-3-8b-r-imat-IQ3_M.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d98e52a724f9ef2acd8688cc7825a5b776f789468be177bf808c9c12b5f04687
|
3 |
+
size 3784823392
|
sfr-iterative-dpo-llama-3-8b-r-imat-IQ3_S.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:1cd74ab4fa8673330c6abaa8386d530c1f25f5d695252d73d06de38ec4b57835
|
3 |
+
size 3682325088
|
sfr-iterative-dpo-llama-3-8b-r-imat-IQ3_XS.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d08930c8b3cdc29c30f636f089b67739dda5c4bdb149333de4aa19fcc1afb5aa
|
3 |
+
size 3518747232
|
sfr-iterative-dpo-llama-3-8b-r-imat-IQ3_XXS.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:763c30bfe2c434917057c66f56ac51b9dbd5515e5b09ef445eea6f475af6702e
|
3 |
+
size 3274912352
|
sfr-iterative-dpo-llama-3-8b-r-imat-IQ4_NL.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f95ab77f9ab096ced6ff03525f9fc70087c90b5fb66d11bddc471b7cb2fc50df
|
3 |
+
size 4677988960
|
sfr-iterative-dpo-llama-3-8b-r-imat-IQ4_XS.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:05d1a19f0ef23540e53356d4c1476194b433c53f798155bb2c27becb87139d6d
|
3 |
+
size 4447662688
|
sfr-iterative-dpo-llama-3-8b-r-imat-Q4_0.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:7cc7b1459aa4a706752d3146e5d900e68d7af692960b7741abdb624985a9643d
|
3 |
+
size 4675891808
|
sfr-iterative-dpo-llama-3-8b-r-imat-Q4_K_M.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:8d3d68484fcfd893799e5f6776f4c7c6ef902fb3728f106dba468c9e5c61277b
|
3 |
+
size 4920734304
|
sfr-iterative-dpo-llama-3-8b-r-imat-Q4_K_S.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:0704dae075b19ef21d5d9ca870ab1e3a12e973aa71bbc05433aa464f7537320b
|
3 |
+
size 4692669024
|
sfr-iterative-dpo-llama-3-8b-r-imat-Q5_K_M.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:2c2ae93d7b565aa96008986a419bf7c6456c1e31d92736f2961d7630420d90a9
|
3 |
+
size 5732987488
|
sfr-iterative-dpo-llama-3-8b-r-imat-Q5_K_S.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:30e443649a761f1bcbb0eb50a8dfb93a09d32bcb099a102920f4f5b34aab7915
|
3 |
+
size 5599294048
|
sfr-iterative-dpo-llama-3-8b-r-imat-Q6_K.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:393af1163456e7176163de61ed25bcba2a101be796bb0f31c9bdbf4c29bec421
|
3 |
+
size 6596006496
|
sfr-iterative-dpo-llama-3-8b-r-imat-Q8_0.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:61e322d34bf50ded57638458bd051424b5591fd51d5d7942e2b7bc90867646d6
|
3 |
+
size 8540770912
|